• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.3
• /
• pp.374-380
• /
• 2022

This paper shows the results of a study that predicts changes in seawater conditions in sea farms using machine learning-based long short term memory (LSTM) algorithms. Hardware was implemented using dissolved oxygen, salinity, nitrogen ion concentration, and water temperature measurement sensors to collect seawater condition information from sea farms, and transferred to a cloud-based Firebase database using LoRa communication. Using the developed hardware, seawater condition information around fish farms in Tongyeong and Geoje was collected, and LSTM algorithms were applied to learning results using these actual datasets to obtain predictive results showing 87% accuracy. Flask and REST APIs were used to provide users with predictive results for each of the four parameters, including dissolved oxygen. These predictive results are expected to help fishermen reduce significant damage caused by fish group death by providing changes in sea conditions in advance.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.178-180
• /
• 2021

In this paper, we design a system to monitor environmental data in fish farms in real-time and provide machine learning-based prediction services to prevent damage on fish farms caused by changes in the sea environment. The proposed system will install an IoT device module consisting of sensors that can measure hydrogen concentration, salinity, dissolved oxygen, and water temperature, which can be transferred to Cloud DB using LTE or LoRa communication technology and then monitor the real-time condition through a web or mobile application. In addition, it has a function to prepare for changes within the environment of fish farms by applying machine learning-based prediction technology using collected data.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2010.04a
• /
• pp.26-27
• /
• 2010

Illegal aquaculture farms that cause marine pollution and ship accidents are managed and controlled by local organizations, but they experience many difficulties due to the vast waters and insufficient human resources. This study aims to utilize new IP-RFID technology to propose an efficient aquaculture farm facility management system This system attaches Ip-Tags to aquaculture farms that can provide facility information and location in real-time through IP communications. This allows for efficient management of facilities and the restriction of illegal farms. This system can also provide locations of farm facilities that sink to the ocean floor, reducing marine pollution caused by lost farms.

• Annual Conference of KIPS
• /
• 2015.10a
• /
• pp.351-354
• /
• 2015

지구 전체 표면적의 약 70%인 바다는 석유를 포함한 각종 수산자원이 풍부하지만 인간은 바다로 접근하기 위해 파도, 태풍 등의 날씨에 절대적인 영향을 받기 때문에 쉽게 접근하기 어렵다. 이 경우 해양 관련 정보를 얻고 분석 및 활용하기 위해 IoT (Internet of Things)의 기반 기술인 센서네트워크를 사용할 수 있다. 하지만 바다에 센서네트워크를 적용하기 위해서는 파도, 태풍을 포함한 염분 등을 충분히 고려해야 한다. 게다가 수중 통신을 사용할 경우 수중에서는 전파를 사용할 수 없기 때문에 음파와 같이 수중에서 통신이 가능한 방법을 선택해야 한다. 따라서 본 논문에서는 해양센서네트워크 기술의 현장 적용을 위한 고려사항에 대해 논의하고, 실제 가두리 양식장에 설치 운용한 사례를 소개한다.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.14 no.2
• /
• pp.147-161
• /
• 2011

Korea is surrounded by sea on three sides. This country has been supplied with a variety of aquaculture products cultivated on shores. There have recently been a lot of studies to have better understanding of the correct location and quantity of aquaculture farms for seaweed, ear shells and fish that cover a wide area of sea. And it is necessary to use the geographic information system and remote sensing to detect the aquaculture farms in order to effectively manage them. This study uses higher resolution aerial images(25 centimeters) than satellite images of 2~2.5-meter resolution that have been ever used, to conduct an accuracy detection of aquaculture farming facilities. It chooses as the case study area the Wando region that has aquaculture farms for seaweed, ear shells and fish. Aerial photos of the island were obtained in this study and an image correction of them was conducted. A spatial database was then constructed in this study and the detection of aquaculture farming facilities was performed. An analysis of facilities inside and outside the permitted areas reveals that there has been an increase in the facilities of seaweed and ear shell aquaculture farms outside the permitted areas. And also it tells that because the facilities of fish aquaculture farms have turned into those of ear shell aquaculture farms, there has been a decrease in permitted facilities, facilities detected on the basis of aerial images, and facilities outside the permitted area. It will be necessary to continuously control and manage the unpermitted facilities, regarding the increase in the facilities inside and outside the permitted area for seaweed and ear shell aquaculture farms. Because the facilities of aquaculture farms cover a wide range of areas(sea) in this manner, it is more effective to depend on high resolution aerial images than a field survey to detect and calculate the facilities. This study comes up with a plan for using aerial images to detect the location and the quantity of the fish aquaculture facilities and then effectively manage them.

• Annual Conference of KIPS
• /
• 2012.04a
• /
• pp.1010-1011
• /
• 2012

적조란 유해조류의 일시적인 대 번식으로 바다를 적색으로 변화시키며 연안 환경 및 바다 생태계에 악영향뿐만 아니라 양식장의 어패류를 집단 폐사 시키는 현상이다. 적조에 의한 양식어업의 피해는 매년 발생하고 있으며 매년 적조방제에 많은 비용을 소비하고 있다. 이 때문에 적조 발생을 미리 예측할 수 있으면 적조에 대한 피해 및 방재 비용을 최소화 시킬 수 있다. 본 논문은 앙상블 학습은 이용한 적조발생 예측 방법을 제안한다. 제안방법은 앙상블 학습의 bagging과 boosting 방법을 이용하여서 적조를 예측의 성능을 향상시킨다. 실험결과 제안방법은 단일 분류기에 비하여서 더 좋은 적조 발생 예측 성능을 보였다.

• Proceedings of the Korean Society of Coastal and Ocean Engineers Conference
• /
• 1998.09a
• /
• pp.65-71
• /
• 1998

유조선의 빈번한 왕래로 발생되는 원유의 누출사고, 무절제한 해안환경 파괴 및 생활하수, 공장폐수의 방류 문제가 최근 대중매체를 통하여 빠짐없이 보도되고 있으며 또 그로 인하여 양식장, 바다낚시, 해수욕장 등 생산 및 생활 환경이 오염되어 일반인들도 직접 간접으로 적지 않게 피해를 입게 되므로 그에 대한 관심이 날로 증가되고 있다. 동해 어항의 경우는 수산물 가공폐수 외에 주변에 주거지 및 상가(횟집) 등이 위치하여 생활오수 등이 다량 발생하고 있다. (중략)

• The Sea:JOURNAL OF THE KOREAN SOCIETY OF OCEANOGRAPHY
• /
• v.9 no.1
• /
• pp.30-39
• /
• 2004

In order to understand the environmental impact of marine cage fish farms, we measured the vertical fluxes of particulate to the sediment, the distribution of organic carbon in core samples, sediment oxygen consumption rate (SOD), and macrobenthos with increasing distance from a fish cage in Miruk island located in Tongyong. The experiment was performed in August 2003. Measured values gradually decreased with distance, indicating that the organic matter in the sediment derived from the fish farm. The dominant macrobenthos species were Tharyx mulifilis, Lumbrineris longifolia, Sigambra tentaculata, and Capitella capitata, occupying 88% of the total population. Capirella capirata, an opportunistic polychaete species, were especially abundant between 0 to 5 m radius range. The estimated impact regions of organic matter enrichment based on sediment consilmption rates and compositions of macrobenthos were in good agreement. Most organic matter derived from the fish farm was deposited within a 10 m radius and then dispersed horizontally to nearby (at least 50 m) surface sediment. The vertical organic carbon fluxes to the sediment at the fish farm were higher by a factor of two than those outside the area. The remineralization organic carbon in the upper sediment layer was estimated to be 50% (1.07 g C m$^{-2}$ day$^{-1}$ ) at the fish farm. In contrast, outside the area, 30% (0.30 g C m$^{-2}$ day$^{-1}$ ) of organic carbon was recycled and the remaining 70% was deposited to the deep sediment layer.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.7 no.3
• /
• pp.145-162
• /
• 1974

시판되는 굴, 담치, 백합 등 패류는 깨끗한 수역에서 생산되어야 한다는 것은 공중보건상 매우 중요하다. 따라서 이들 서식해역에 대한 위생상태를 평가하는 것은 패류위생관리에 있어 가장 기본적인 일이다. 본 조사는 미국 Alabama주 West Fowl River 지역에 있어 패류위생관리상 양식장의 등급분류와 관련한 굴(Crassostrea virginica) 및 그 서식수의 위생학적 성상을 알기 위하여 실시되었다. 이 연구는 미국 국제개발처(AID) 연수계획에 의하여 이루워 졌음을 밝혀 둔다. 조사지역에 있어서의 감염원은 가옥, 소업체, 가축, 물새, 야생동물 등에 의한 일반적인 발기물과 배설물로 되어 있고 특별한 하수나 공업배출물은 없었다. 이 조사는 이 지방에 있어 건조기인 10월에 실시되었으며 조사기간중의 총 강우양은 0.08mm, 평균 기온은 $21.3^{\circ}C$, 바람은 동풍이 우세하였다. 저질은 뻘로 되어 있고 저호시 평균수침은 약 1m이며 조사기관중 평균 수온은 $26.0^{\circ}C$였다. 염분은 조사지점에 따라 상이하였으며 바다로 향한 강의 흐름에 따라 증가하고 평균 염분은 $18.0\~28.7\%$의 범위에 있었다. 해수 및 굴 양자의 Coliform 함양은 Fecal coliform 보다 시종 높았으며 Coliform 및 Fecal coliform의 수는 바다로 향한 강의 흐름에 따라 점차 감소하였다. 이것은 해수의 오염은 주로 지표수의 유입에 의한다는 것을 말하고 있다. 굴의 Coliform 및 Fecal coliform MPN은 해수의 그것보다 $11\~12$배 높았다. 조사수역에 설정한 7개소 Sampling station 중 오직 Station 7만이 패류양식장 허가해역의 세균학적 수질기준에 합당하였으며 굴의 Fecal coliform MPN은 시판용 패류의 세균학적 기준을 초과하고 있었다. EC test 양성시험관에서 분리된 세균의 $97.6\%$가 E. coli group로 나타나 EC test는 Echerichie Coli 시험에 매우 효과적인 방법임을 알 수 있다.

• The Korean Journal of Malacology
• /
• v.16 no.1_2
• /
• pp.49-54
• /
• 2000

The growth of short neck calms, Tapes philippinarum was compared between two areas, Songdo in Kamak Bay and Sudo in Chinhae Bay from April 1999 to March 2000. The short neck clams in Kamak bay grew from 13.84${\pm}$0.04 mm to 35.43${\pm}$1.22 mm in shell length, while those in Chinhae bay grew from 13.84${\pm}$0.04 mm to 31.91 ${\pm}$1.72 mm in 12 months. Shell height, shell breadth and total weight of the short neck clams in Kamak bay were also greater than those from Chinhae bay. Bottom quality of Chinhae bay showed negative skewness, and that of Kamak bay was positive skewness of Chinhae bay seems to be caused by the effect of strong tidal current. This may indicate that Kamak bay is better area than Chinhae bay for the culture of the short neck clams.